"Checksum mismatch -- will transfer entire file"

[Victor Sudakov]

Dan Nelson wrote:
> In the last episode (Dec 25), Victor Sudakov said:
> > > I cvsup the FreeBSD CVS repository daily from cvsup.ru.freebsd.org. 
> > > Both the client and the server run CVSup Software version: SNAP_16_1h,
> > > Protocol version: 17.0.
> > > 
> > > Recently I noticed that there are lots of messages "Checksum mismatch --
> > > will transfer entire file" about all kinds of downloaded files.
> > > 
> > > What could be the reason? Is my CVS repository corrupt or what? Is there
> > > a way to check the integrity of the entiry repository?
> > > 
> > > I have read about there being a "checksum mismatch" problem in CVSup
> > > version before 15.4, but I am using SNAP_16_1h already.
> > > 
> > > If this question is offtopic here, please direct me to a more relevant
> > > mailing list.  TIA.
> >
> > Am I the only one to have this problem?
> 
> I see this too.  Running cvsup -k and looking at the "bad" files shows that
> the differences seem to be commit dates before 2000; one end has them as
> "99", and the other has "1999", which causes the checksum to change.

This is interesting, but I observe different issues. Consider for
example the following diff:


*** #cvs.cvsup-20614.43	2009-12-29 08:56:31.000000000 +0600
--- cpucontrol.c,v	2009-12-29 04:58:12.000000000 +0600
***************
*** 36,41 ****
--- 36,52 ----
  branches;
  next	;
  
+ 1.3.2.1
+ date	2009.08.03.08.13.06;	author kensmith;	state Exp;
+ branches
+ 	1.3.2.1.2.1;
+ next	;
+ 
+ 1.3.2.1.2.1
+ date	2009.10.25.01.10.29;	author kensmith;	state Exp;
+ branches;
+ next	;
+ 
  1.2.2.1
  date	2009.01.12.15.48.22;	author stas;	state Exp;
  branches
***************
*** 52,68 ****
  branches;
  next	;
  
- 1.3.2.1
- date	2009.08.03.08.13.06;	author kensmith;	state Exp;
- branches
- 	1.3.2.1.2.1;
- next	;
- 
- 1.3.2.1.2.1
- date	2009.10.25.01.10.29;	author kensmith;	state Exp;
- branches;
- next	;
- 
  
  desc
  @@
--- 63,68 ----
***************
*** 815,817 ****
--- 815,818 ----
  a233 1
  		WARNX(0, "error opening %s for reading", dev);
  @
+ 

We can observe the following:

1. There is an extra newline near the end of the file.

2. The changes in the file are the same but placed in different
positions in the file.

Any ideas?

-- 
Victor Sudakov,  VAS4-RIPE, VAS47-RIPN
sip:sudakov at sibptus.tomsk.ru